Table 12. Comparisons with recent related studies based on Dataset I.

Article	Method	Performance metrics
		Accuracy (%)	Sensitivity	Precision	Specificity	F1 score
Ahmad et al. (2017)	AlexNet	75.4	–	–	–	–
Agrawal et al. (2017)	GF1+Inception V3+VGG+SVM	96.1	0.852	0.847	0.978	0.827
Pogorelov et al. (2017)	ResNet+LMT2	95.7	0.826	0.829	0.975	0.802
Pogorelov et al. (2017)	GF+Decision Tree	93.7	0.748	0.748	0.964	0.747
Pogorelov et al. (2017)	GF+Random Forest	93.3	0.732	0.732	0.962	0.727
Nadeem et al. (2018)	GF+LBP3+Haralick+LR4	94.2	0.774	0.767	0.966	0.707
Thambawita et al. (2018)	GF+CNN	95.8	0.958	0.9587	0.9971	0.9580
Owais et al. (2019)	ResNet+DenseNet+MLP5	92.5	0.993	0.946	–	0.934
Gastro-CADx		97.3	0.9715	0.9718	0.9959	0.9715

Notes:

¹Global features.

²Logistic model tree.

³Local binary pattern.

⁴Logistic regression.

⁵Multilayer perceptron.